Examining the runes
Neil Merrick looks at the different ways in which a school and an FE college chose to measure their success.
Exam results are the key performance indicators used at North Westminster Community School. For the past two years, this inner-London school has compared students' A-level and GCSE results to find out whether they have done better or worse than expected in the sixth form.
The school uses a formula suggested by the Audit Commission in its 1993 report, Unfinished Business. Each student is given a GCSE score based on the grades they achieved in English, maths and their best five other subjects. An A grade is worth seven points, a B grade six points and so on down to one point for a G grade.
Through multiplying this figure by 1.05 and subtracting 29.46, the school is able to calculate the total number of UCAS points a student would be expected to gain at A and AS-level. Students who do better than expected get a positive residual score while those who do worse get a minus. By adding up all of the residual scores, North Westminster can assess the total value which has been added to sixth-formers' results.
Jerry Hicken, the school's head of assessment, has used the formula to produce retrospective scores for GCSE students since 1990 and for A-level students in 1992. He can therefore assess how sixth-formers performed during each of the past three years.
In 1992, the first year for which both GCSE and A-level data was available, the average residual score was 0.03, suggesting that students had performed more or less as expected. In 1993 the figure was -1.10, suggesting students did less well, while this summer it rose to 1.37. The average residual score hides vast variations in the results achieved by individuals. Nevertheless, the study shows that in 1994 large numbers of students did better than the formula predicted. An average residual score of two points would be equivalent to each candidate achieving one grade higher than expected in one A-level.
Unfinished Business added that students with a GCSE score of 32 or below should not be admitted to A-level courses without them first being made aware of their low chances of success.
The research at North Westminster confirms that students who score 32 in their GCSE exams frequently gain only modest A-levels. Students whose GCSE scores are between 32 and 36 are categorised as "at risk" and observed carefully during the early weeks of an A-level course.
"We like to see whether they are coping or not. The likelihood is that they will be less hard-working," said Mr Hicken. "There is a limit to the time staff can spend monitoring students, so it's better to be aware which ones are most likely to need monitoring".
Only a minority of students follow vocational courses at intermediate or advanced level, so the school has concentrated on comparing GCSE and A-level scores. It also studies GCSE results according to the gender and ethnic background of students and compares external exam grades with test results pupils achieved when they were admitted to the school at 11. All exam and test scores are recorded on a computer which can be easily accessed by school staff. Mr Hicken, however, refuses to get too carried away by performance indicators. "You should always be careful when you are using quantitative measures and not treat figures more seriously than they should be treated". he said. "You must look at qualitative measures as well".